Massive amounts of data can now be obtained with modern equipment, with some spectrometers generating thousands of spectra in a few minutes. While this allows one to probe samples with large variability with unprecedented performance, it poses the problem of data processing. In this presentation, an overview is provided of a computational system implementing information visualization methods, which can be used in various analytical tools. Examples will be given of the combination of multidimensional projections, parallel coordinates and machine learning techniques to enhance the sensitivity of biosensors whose principle of detection was impedance spectroscopy. In such biosensors, antigens specific to antibodies for tropical diseases were immobilized in layer-by-layer (LbL) films adsorbed on interdigitated gold electrodes. Small differences in the impedance spectra could be used as fingerprint for distinct diseases. In another example, real-time analysis of surface-enhanced Raman spectroscopy (SERS) data will be shown, where single molecule detection was reached.

Information visualization methods for enhancing spectral analysis

245th National Meeting (2013)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.